Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Value of Ferrite bead

FreshmanNewbie

Full Member level 6
Joined
May 10, 2020
Messages
374
Helped
0
Reputation
0
Reaction score
2
Trophy points
18
Activity points
3,725
I have seen a couple of ethernet PHYs whose manufacturer datasheet recommends placing a ferrite bead on the power supply input of the PHY device who's value is, "Generally, a 100-220Ω (at 100 MHz) ferrite bead is used.". Mostly it is on the analog domain of the PHY.

You can find it in the first point under Power Section on page 1 of this Checklist : https://ww1.microchip.com/downloads/en/DeviceDoc/KSZ8041MLL-Hardware-Design-Checklist-00002856A.pdf

Can someone tell me why is such a value chosen at that particular frequency of 100MHz? Any specific reason with respect to the operating frequencies of Ethernet?
 
Answer is very simple. Ferrite beads are generally specified with impedance at 100 MHz, except for low frequency power ferrites.
 
Hi,

you can see there are capacitors on both sidees of the bead.
Let´s assume the 22uF isn´t suitable fro 100MHz .. thus let´s ignore it.

So we have a 100nF at both sides. It´s impedance at 100MHz is just 0.016 Ohms. (ignoring stray inductance)
Thus we have a 100Ohms : 0.016 Ohms ratio to suppress 100MHz.

Without the capacitors the bead is in best case useless, but usually counter productive because it makes the power supply weak. If you use the bead you always have to use capacitors, too. Also for the math.

Klaus
 
Hi,

you can see there are capacitors on both sidees of the bead.
Let´s assume the 22uF isn´t suitable fro 100MHz .. thus let´s ignore it.

So we have a 100nF at both sides. It´s impedance at 100MHz is just 0.016 Ohms. (ignoring stray inductance)
Thus we have a 100Ohms : 0.016 Ohms ratio to suppress 100MHz.

Without the capacitors the bead is in best case useless, but usually counter productive because it makes the power supply weak. If you use the bead you always have to use capacitors, too. Also for the math.

Klaus
Thank you for the answer! Is the value of the bead (100 to 200 ohms) or the frequency, 100MHz, is anyway related to the ethernet operation?

Just asking whether this recommendation of the values is specific to the ethernet?

Or can this value of the ferrite bead can be used for say other interfaces, if applicable, like say, USB?
 
Hi,

FvM explained it already. 100MHz was chosen as general frequency, so you can do a fast and easy selection between ferrites. If you need more information regarding frequency, impedance, resistance, maximum current .. you still need to read the datasheet.

This 100MHz is not related to Ethernet only.

The bead isn´t in the signal path at all, it´s in the power supply path, in combination with a bypass capacitor.

Klaus
 
Hi,

FvM explained it already. 100MHz was chosen as general frequency, so you can do a fast and easy selection between ferrites. If you need more information regarding frequency, impedance, resistance, maximum current .. you still need to read the datasheet.

This 100MHz is not related to Ethernet only.

The bead isn´t in the signal path at all, it´s in the power supply path, in combination with a bypass capacitor.

Klaus
Thank you. If it is not specific to ethernet, why use a ferrite bead at all? Can it be ignored simply for this?
 
Hi,

A ferrite bead is generally used for HF noise suppression.

Functionally it may be ignored (depending on application) but leagally your device needs to comply with EMI/EMC standards.

On some applications - especially when analog signals are involved - omitting the bead may degrade performance.
There is a good reason why the IC has split analog and digital supply.

Just accept it.

Klaus
 
Hi,

A ferrite bead is generally used for HF noise suppression.

Functionally it may be ignored (depending on application) but leagally your device needs to comply with EMI/EMC standards.

On some applications - especially when analog signals are involved - omitting the bead may degrade performance.
There is a good reason why the IC has split analog and digital supply.

Just accept it.

Klaus
Thank you for the answer.

In one of your earlier responses to my question, you mentioned a important point which actually answered my question.

You mentioned that the ferrite bead is not placed in the signal line but in the power line.

Hence, it doesn't matter if the interface is Ethernet, USB or PCIe or anything. But in that case, why bother to install ferrite bead at all? What does it actually help to do in the power line?

Does it help to reduce some sort of conducted emissions peaks?
 
The ferrite material is meant to be lossy at HF, so the power
rail noise goes to heat rather than ringing up whatever tank
has been created. That's the "plan" anyhow. If cut-and-try
voodoo is something you'd call a plan.
 

LaTeX Commands Quick-Menu:

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top